AverageMarcus/blog
1
0
Fork 0
Code Issues Pull Requests Releases Activity

Fixed broken image

Browse Source
This commit is contained in:
Marcus Noble 2017-06-07 06:25:01 +01:00
parent 5f7ddb4c30
commit 2a30a70596
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/posts/2016-10-25-hackathonarama-2.markdown
View File

@ -20,7 +20,7 @@ The afterparty, Chillference, was a really good way to chat with lots of people
Last year I [posted some tips](/2015-10-06-hackathonarama/) on surviving an all-night hackathon. I still stand by these tips and encourage people to read them if they plan to attend such an event. This year I went in to it much more prepared. I knew what to expect this time, I had an idea of what my limits would be, I actually had some ideas and I wasn't alone. Going along with someone ([Dan](https://twitter.com/danielthepope) :wave:) made things _much_ less daunting than last year.
{{> picture alt="Hackference 2016" caption="Everyone ready to hack" url="images/hackference2016.jpg" }}
{{> picture alt="Hackference 2016" caption="Everyone ready to hack" url="/images/hackference2016.jpg" }}
Even though I had some idea of what I wanted to do I started out hacking at something else to start with. Initially I planned to create an app that would be able to determine the reading age of a user's tweets. This turned out to not be so feasible, mainly due to the brevity of tweets. Instead I made use of the [sentiment analysis API](https://westus.dev.cognitive.microsoft.com/docs/services/TextAnalytics.V2.0/operations/56f30ceeeda5650db055a3c9), part of Microsoft's [Cognitive Services](https://www.microsoft.com/cognitive-services). This ended up being pretty fun and surprisingly simple. All you need to do is send it some text and it'll return a value between 0 and 1 indicating what it belives the sentiment of the text is. I used this to take a look at the current political hopefuls in the US, [Donald](http://tweetanalysis.marcusnoble.co.uk/user/realDonaldTrump) and [Hillary](http://tweetanalysis.marcusnoble.co.uk/user/hillaryclinton), both who (at time of writing) see pretty positive. It'll be interesting to see how that changes after the election. Take a look for yourself: [Tweet Analysis](http://tweetanalysis.marcusnoble.co.uk/) ([Code](https://github.com/AverageMarcus/TweetAnalysis))